Skip to content

Independent marketplace for verified cloud accounts

Hetzner vs Vultr

This comparison comes down to one trade: Vultr puts compute in many more places around the world, while Hetzner delivers more performance per unit of spend from a narrower footprint.

Buy Hetzner Account
Germany, EU hardware value

Hetzner

Regions noted in source
Germany, Finland, United States, Singapore
Configurations listed
3
From$50 View account
Buy Vultr Account
Wide data center footprint

Vultr

Regions noted in source
Every populated continent
Configurations listed
3
From$25 View account

There is a version of this comparison that lists features, finds them broadly similar, and tells you to pick whichever is cheaper. That version is useless, because the two providers are not really competing on the same axis. Hetzner competes on how much machine you get. Vultr competes on how many places you can put one.

Everything else follows from that. Once you know where your users are and how tight your latency budget is, the decision usually makes itself.

The trade, stated plainly

Hetzner concentrates capacity in a small number of locations it owns and operates, mostly in Europe, with a presence in the United States. Dense, efficient, cheap per core and per gigabyte of RAM.

Vultr runs a much longer list of locations spread across North America, South America, Europe, Africa, the Middle East, Asia and Australia. Running a server close to users in Sao Paulo, Johannesburg, Bangalore or Sydney is a one-region decision rather than a second-provider project.

You are choosing between depth in a few places and presence in many. Neither is a better engineering choice in the abstract.

When footprint is the whole answer

Latency is not a preference, it is a property of distance and routing. If your application is chatty, if it does several round trips to render a page, or if it is interactive in a way users feel, then a server on the wrong continent is not fixable with a bigger instance.

Three workload shapes where Vultr’s spread decides it:

  • Multiplayer and real-time applications where tens of milliseconds change the experience.
  • Regional compliance or performance requirements that need compute inside a specific country, not merely a nearby one.
  • Products with a genuinely distributed user base that need origin servers in several regions rather than a CDN in front of one.

Caching and a CDN solve static delivery. They do not solve a write path that has to reach your database. If your users in Asia are writing data, the write is going wherever your primary lives.

When price to performance is the whole answer

If your traffic is European, or if your workload is asynchronous and nobody is sitting watching it, location count stops being interesting and cost per unit of work takes over.

Hetzner’s advantage here is structural. The company owns its facilities, runs them at high density and sells on thin margin. For memory-heavy workloads in particular, the gap in what you get for a given monthly spend is large enough to change architecture decisions: you can keep a dataset in RAM that you would otherwise have to shard, or run a build fleet you would otherwise have to queue.

Typical cases where this wins: CI runners, data processing, self-hosted observability stacks, staging environments that need to look like production, large caches, and anything where the honest answer to where should this run is wherever it is cheapest.

Configuration flexibility and instance choice

Vultr offers more shapes of machine. Shared and dedicated vCPU lines, high frequency compute aimed at single-threaded performance, bare metal, and GPU instances sit in the same console under the same API. If your workload mix varies, that breadth means fewer compromises and no second account.

Hetzner’s catalogue is narrower by design: shared and dedicated vCPU cloud instances, plus a separate and well-regarded dedicated server business with auction-style availability. There is no GPU line in the cloud product, which rules Hetzner out for training and most inference work. If that is your workload, GPU instances for training and inference explains what to look for before you commit to any provider.

What the invoice looks like

Both providers are predictable compared with usage-billed hyperscalers. Instances bill on a clear rate, bandwidth comes with an allowance rather than being metered from zero, and neither platform has many ways to quietly accumulate charges.

Two things still catch people. The first is bandwidth overage once you exceed the included allowance, which is where high-traffic media and download-heavy workloads get expensive on any provider. The second is forgotten resources: snapshots, detached volumes, reserved addresses and idle instances in a region you spun up for a test. Neither provider will clean those up for you. Why cloud bills surprise people covers the pattern, and it applies to cheap providers more often than you would expect, precisely because nobody is watching a small invoice.

Operational feel

Vultr’s console and API are closer to the DigitalOcean model: approachable, quick to provision, with a marketplace of prebuilt application images and a reasonable amount of platform tooling around the VM. It is a comfortable environment for a developer who wants to be running in five minutes.

Hetzner’s console is clean and plain, and the API is solid. The experience assumes you know what you want. The dedicated server side in particular is a hosting product with hosting conventions, not a cloud abstraction, and that surprises people arriving from a pure cloud background.

Both have good Terraform providers, which in practice matters more than either console. If your infrastructure is described in code, the console is a dashboard, not a workflow.

Storage, snapshots and moving between regions

Block storage exists on both platforms and behaves the way you would expect: create a volume, attach it to an instance in the same location, resize it later. The practical differences show up at the edges rather than in the headline product.

Volumes are location-bound on both providers, which means a region move is a data move, not a detach and reattach. On Vultr, where the temptation to spread across many locations is the whole point, that constraint bites more often. Plan the data layer first and the compute placement second, or you will end up with servers near your users and a database three regions away, which is worse than putting everything in one place.

Snapshots are the other recurring issue. They are cheap individually and invisible in aggregate, and on both platforms they persist until someone deletes them. Build snapshot expiry into whatever automation creates them rather than trusting a quarterly cleanup that will not happen.

For object storage, Vultr offers an S3-compatible service alongside compute. Hetzner’s storage products are oriented differently, toward network-attached storage boxes rather than a general object API, so an application written against S3 semantics usually keeps a separate object store. If you are not sure which storage model your workload actually needs, cloud storage types explained is the shorter route than discovering it in production.

Picking by deployment shape

Deployment shape Lean toward Reason
Single-region European product Hetzner Best value where the users already are
Users on four or more continents Vultr Origin servers near users without a second provider
Batch processing and build fleets Hetzner Latency-tolerant work rewards cheap cores and RAM
GPU or high frequency single-thread needs Vultr Those instance families exist in the catalogue
Compliance tied to a specific country Vultr Longer list of national locations
Large in-memory datasets Hetzner Memory per unit of spend changes the design

Running both

Plenty of teams do not choose. They put the latency-sensitive edge of the application on Vultr in several regions and keep the heavy, cost-dominated work on Hetzner, linking the two over a VPN or private tunnel. That is a reasonable design, but it is two providers, two billing relationships, two sets of credentials and two failure domains to understand. Read when multi-cloud helps before you commit, because the answer is less often than people assume.

Accounts and getting started

Both providers verify new signups and both have become stricter about payment methods and billing regions. If a cold signup is blocked or you need capacity today, an established account is one way through. The listings for Buy Hetzner Account and Buy Vultr Account set out what each includes.

TrustCloudHub is an independent marketplace and is not affiliated with Hetzner or Vultr. Where credit is attached to an account, that credit belongs to the provider’s programme: it expires, it carries service restrictions, and the remaining balance should be checked inside the provider console after transfer rather than assumed from a listing. If you want to see how the same value question looks against another nearby competitor, Digital Ocean vs Vultr approaches it from the ecosystem side instead.

Other comparisons to consider

Hetzner vs Linode
vs

Hetzner vs Linode

Both platforms were built by engineers for engineers, but they answer to different owners now. This compares operating philosophy and…

Hetzner vs Digital Ocean
vs

Hetzner vs Digital Ocean

Hetzner sells hardware value from a mostly European footprint. DigitalOcean sells a developer platform with managed add-ons and deep documentation.…

AWS vs Hetzner
vs

AWS vs Hetzner

For steady always-on compute the price gap between Hetzner and AWS is wide. This comparison works out what that gap…

Google Cloud vs Vultr
vs

Google Cloud vs Vultr

Both platforms will rent you a GPU, and that is where the similarity ends. Vultr gives you the hardware; Google…

Background reading

Reserved, Spot and On-Demand Pricing Compared
Cost Management 7 min read

Reserved, Spot and On-Demand Pricing Compared

The same machine sells three ways at three prices. What changes is the commitment you make and the interruption risk you accept.…

Decided which platform fits?

Both accounts are documented in the catalog, with configurations, credit balances and service limits listed before you buy.

Telegram